Appointment scheduling is prevalent in various everyday problems. Consider for example a doctor serving his patients, a courier delivering parcels, or a service man visiting his clients. In all of these problems, it is desired to provide a high quality of service. In particular, this means creating a schedule that satisfies the wishes of both parties involved, i.e., the clients and the service provider.
Several techniques have been developed to determine the optimal schedule given the clients and schedule characteristics. My research focuses on improving the schedules by using operations research and machine learning. In particular, an approach will be developed with which the schedules can be repeatedly updated.
Secondly, we will develop frameworks that also fit into a delivery context. One major complication here is that it might not always be possible to plan the order of arrivals; predictions must also be made. A partnership with the Dutch mail corporation PostNL allows us to test our results, and improve their parcel delivery service.